Hardware Virtualized Flexible Network for Wireless Data-Center Optical Interconnects [Invited]

Abstract

Future networks are set to provide a wide range of services over a multitude of technologies carrying a broad range of traffic profiles. The emerging mobile cloud computing applications are imposing numerous challenges on network operators to enable ubiquitous and on-demand service provisioning. To cope with the rising demands wired and wireless networks, together with IT infrastructures, need to be part of a converged platform. They should allow end-to-end network and cloud composition and virtualization, which are required for ubiquitous and distributed access. Flexible and programmable optical networks can integrate and merge IT and wireless technologies effectively and efficiently. In view of this, we propose the use of a time shared optical network (TSON) for the interconnection of wireless access networks and IT resources and expand its functionality and dynamicity for integration in a converged yet heterogeneous environment. TSON is also extended to facilitate a virtualization-enabled multitechnology echo system. The extensions aim at facilitating hardware virtualization of the TSON technology and include support of higher rates, flexible subwavelength or wavelength modes of operation, and dynamic memory allocation facilitating infrastructure virtualization and reprogramming. An experimental implementation evaluates the impact of TSON's variable aggregation buffer memory requirements on different frame sizes and its effect on network bandwidth and latency. Using experimental data, simulation studies introduce and investigate large-scale scenarios of hardware virtualization and investigate the resource efficiency advantages enabled by the proposed solution. A theoretical evaluation of the network performance of the proposed TSON solution is also provided, and results on the delay performance of the converged infrastructure and the associated trade-offs are presented.

Wireless-Optical Network Convergence: Enabling the 5G Architecture to Support Operational and End-User Services

“…The proposed model takes a holistic view, considering jointly mobile FH and BH functions, to ensure appropriate allocation of the required resources across all domains. To identify the best performing FH and BH VIs, detailed power consumption and end-to-end delay models, describing the optical and wireless network as well as the compute domains and the associated interfaces, are considered [9]. The results obtained using this modeling framework, focus on the specific use case of joint FH and BH optimisation, assuming delivery of CDN services to the endusers.…”

“…, Markos Anastasopoulos (1) , Dimitra Simeonidou (1) , Ignacio Berberana (2) , Dimitris Syrivelis (3) , Paris Flegkas (3) ,Thanasis Korakis (3) , Daniel Camps Mur (4) , Ilker Demirkol (5) , Jesús Gutiérrez (6) , Eckhard Grass (6) , Qing Wei (7) , Emmanouil Pateromichelakis (7) , Nikola Vucic (7) , Albrecht Fehske (8) , Michael Grieger (8) , Michael Eiselt (9) , Jens Bartelt (10) , Gerhard Fettweis (10) , George Lyberopoulos (11) , Eleni Theodoropoulou (11) Initiative (OBSAI) and the Open Radio Interface (ORI), with CPRI currently being the most widely used standard. FH is responsible to carry the RU wireless signals, typically over an optical transport network, using either digital transmission (e.g.…”

Location-Recommendation-Aware Virtual Network Embedding in Energy-Efficient Optical-Wireless Hybrid Networks Supporting 5G Models

“…Equations (6) and (7) both guarantee that, on the link traversed by a path, there is only one wavelength assigned to this path. Equation (8) ensures that, for each path, the number of wavelengths used for the incoming and outgoing paths are the same at each intermediate substrate node.…”

“…Proof: The authors in [8][9][10] have highlighted the traditional virtual network embedding without the awareness of location recommendations is a NP-hard problem. In addition to the constraints (listed in Eqs.…”

“…Obviously, advertisement targeting has become a classic mobile cloud computing service, and the small-scale DC located close to the end devices of users enables a deeper understanding of user preferences. To collect location histories of users, communication among the smallscale DC and users should be achieved; then, the requirement to interconnect users with DCs introduces the additional need for a 5G wireless network that is seamlessly integrated with the optical DC infrastructure [8][9][10]. Figure 1(b) shows a converged infrastructure integrating optical metro and ubiquitous wireless access technologies in accordance with the 5G model.…”